Friday, 2011-06-03

*** Quits: giallu (~giallu@fedora/giallu) (Ping timeout: 240 seconds)01:23
*** Joins: Cupertino (~Cupez@unaffiliated/cupertino)02:25
*** Quits: wolog (~wolog@wolog.info) (Ping timeout: 240 seconds)02:40
*** Joins: wolog (~wolog@wolog.info)02:41
*** Joins: kirillka (~Miranda@195.242.142.17)03:16
*** Joins: Rixie (~Rixie@188.177.20.182)03:22
*** Quits: Rixie (~Rixie@188.177.20.182) (Quit: Rixie)03:38
*** Joins: cerf (~quassel@mcl71-3-78-241-52-239.fbx.proxad.net)03:38
*** Quits: scribe9343423 (~scribe934@static.96.23.63.178.clients.your-server.de) (Remote host closed the connection)04:28
*** Joins: scribe9343423 (~scribe934@static.96.23.63.178.clients.your-server.de)04:28
*** Joins: paulr (~a@212.85.5.19)04:29
paulr.04:30
*** Joins: giallu (~giallu@fedora/giallu)04:57
*** Joins: asm89 (~asm89@5ED18FE0.cm-7-2c.dynamic.ziggo.nl)04:59
asm89hm, not so busy in here aye? :p04:59
asm89i wondered if there is support for cgit anywhere04:59
asm89for the source integration plugin04:59
asm89if not, ill try to write it, and i'd like to know wheere to submit patches :)05:00
*** Quits: paulr (~a@212.85.5.19) (Ping timeout: 240 seconds)05:48
dhx1paulr: come back! :D06:11
*** Joins: sgimeno (~chatzilla@163.117.206.10)06:24
*** Quits: Cupertino (~Cupez@unaffiliated/cupertino) (Quit: I give up...)06:57
*** Joins: Rixie (~Rixie@188.177.20.182)07:38
*** Joins: wolfman2_ (~wolfman20@rrcs-70-63-208-211.midsouth.biz.rr.com)07:45
asm89hm fixed07:47
asm89will try to find out how to contribute this later07:48
*** Quits: wolfman2000 (~wolfman20@rrcs-70-63-208-211.midsouth.biz.rr.com) (Ping timeout: 252 seconds)07:48
dhx1asm89: jreese is the person to speak to08:10
dhx1jreese: can I suggest we move source-integration to github so other people can easily fork it and add new integration plugins, etc08:11
jreesedhx1: I've been wanting to move all of mantisforge's to github anyways, just not sure how to manage everything08:13
dhx1jreese: in terms of setting up a github repo for each of your plugins... or all core plugins at once, etc?08:14
jreeseboth08:14
jreeseie, we have a "mantisbt" organization for the core repos, but I'm not sure if plugins should share that org, or if we should create a second org, or what08:15
dhx1I guess it depends on how you want to control commit access08:15
jreeseright08:16
*** Joins: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net)08:16
*** Parts: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net) ()08:17
dhx1it's management overhead vs. whether we will need to give commit access to a few plugins at a time to random people08:17
jreeseanywho, asm89, ideally I would like to have a basic git plugin that uses the actual git tools to clone a repo, and do all of the reading locally, rather than scraping a web interface, and then extend that to support linking to external repo viewers like github, gitweb, and cgit08:17
dhx1IMO it might be easier bundling it all within the core repository so that it's easier for core developers to update plugins as the core changes08:17
dhx1agreed... I have been holding off on pushing through HgWeb patches for that reason too08:18
asm89jreese: the code is done09:06
asm89i think the current way of working is actually quite nice09:07
asm89because in this way you do not have to have the git repository on the same server as the git repo itself09:07
jreeseyes, I agree, except that every time the upstream project changes their output format it breaks imports into mantis... =\09:08
jreesenevertheless, if you already have the code done, I'll gladly accept it09:08
asm89its a bit hacky right now09:08
asm89ill clean in up the next couple of days, is that ok?09:09
jreesesure thing09:09
asm89cool :)09:10
asm89first let me add this channel to my autojoin list so i dont forget ;)09:11
asm89jreese: how do you want the "patch" ?09:11
asm89i only created a new SourceCgit plugin09:11
asm89based on the webgit one09:11
asm89i can put it up on github when im done?09:11
jreesepreferably either push a branch to mantisforge or github, or if you don't want to do that, use `git format-patch` 09:11
asm89ok09:12
asm89jreese: it isn't on mantisforge right?09:16
asm89i mean... only on mantisforge09:16
asm89not on github09:16
jreeseit's on both09:17
jreesehttp://git.mantisforge.org/w/source-integration.git or https://github.com/jreese/source-integration09:17
asm89ah :)09:17
asm89it's not listed here: http://git.mantisforge.org/w/source-integration.git09:17
asm89let me fork that ;)09:18
dhx1jreese: thanks for setting up the GitHub repo :)09:19
jreesenp, it's apparently been set up for a while and I forgot about it :P09:19
dhx1well not the project... but at least a repo :)09:19
jreese?09:19
dhx1well it seems like it's a repository under your account, as opposed to a separate project on GitHub09:20
jreeseyeah, I apparently pushed a mirror there at least 5 months ago and forgot about it :P09:20
dhx1yep09:20
dhx1at least it's easy to fork stuff, see other people's contributions, etc on GitHub09:21
jreeseyeah, it's also a hell of a lot easier to manage repos and ssh keys09:21
*** Joins: daryn (~daryn@h158.249.190.173.static.ip.windstream.net)09:21
jreesethat's part of why I would like to move away from mantisforge, having to set up a whole bunch of repo-specific passwords and ssh keys is really obnoxious09:22
dhx1yep09:22
dhx1daryn: hey09:22
* daryn runs and hides09:22
daryn:)09:22
darynhey09:22
darynbeen mighty quiet in here lately09:24
dhx1yeah :(09:24
dhx1are there any plans to use your refactor branch?09:24
daryni got yanked off on two other projects at work09:24
dhx1I was providing some patches to paul's db branch on GitHub09:25
daryni do have plans to continue. trying to finish up the internal mantis project09:25
dhx1then realised I probably should have been working off your branch instead09:25
darynso I can get to it09:25
dhx1yep09:25
dhx1just trying to work out where we should be working from09:25
darynwon't matter much i think09:25
daryni'll take whatever is latest and modify to fit the new layout09:25
asm89some kind of mantisforge, but then not mirroring, just a nice overlay on top of github09:25
asm89would be nice09:25
dhx1ok, so I'd be OK to continue working on paul's database branch?09:26
dhx1these changes are affecting many parts of MantisBT (almost every file)09:26
dhx1because functions are being replaced, code relating to queries rewritten, etc09:27
darynyeah should be fine09:27
dhx1however paulr's code isn't as good as what you had09:27
dhx1should clarify, in terms of OO :)09:27
darynwell, i've got updates that maybe would be better...i haven't pushed in a while09:28
dhx1one of the reasons I haven't done much lately is because I don't want to contribute to a branch that'll be impossible to merge later09:28
darynright09:28
dhx1(due to the wide differences it'll have)09:28
dhx1I suggest creating a branch in the mantisbt project on GitHub for development work we're doing09:29
dhx1not master, more like master-next09:29
daryni was planning to do that once we had a clearer idea of where we're going.  I've refactored refactor branch several times :)09:30
dhx1:)09:30
darynthough I think now i've got a pretty good handle on it09:30
dhx1:)09:30
darynmaybe i should change the name of the branch.09:31
dhx1paul's db branch has had a fair bit of work done to replace database-specific query row count functions with PHP code to perform the same action... or SQL COUNT() queries09:31
dhx1other stuff too09:31
dhx1I plan on doing a lot more there to modernise things (fully parametrised queries for one)09:32
daryni think i had incorporated his latest changes in my branch too but i'd have to check again09:33
jreesedhx1: I thought we already had paramaterised queries with db_query_bound() ?09:34
dhx1April 19th is the latest commit in paul's branch09:34
dhx1jreese: most of the time we do... but not always09:34
dhx1I was going to play with the ADO approach of creating a statement object and calling "addParameter(...)" functions to pass in data09:35
asm89is mantis working on a oo refactor?09:35
jreeseasm89: ever so slowly :P09:35
dhx1asm89: manpower depending... it's more a case of doing things in small steps09:35
darynjreese,  not any slower than any other changes :P09:35
dhx1haha09:36
darynmaybe even faster...you just don't see it yet09:36
asm89hm09:36
asm89nice idea09:36
asm89but09:36
asm89using any frameworks ?09:36
dhx1asm89: not unless they're needed...?09:36
darynasm89, no frameworks. too much overhead for what we're doing09:36
dhx1"frameworks" are the joke of PHP :)09:37
darynwe are sort of following the zend layout though09:37
dhx1if I had to summarise PHP frameworks: "yet another pile of abstract code that helps achieve vague outcomes for hypothetical projects that don't exist" :)09:39
asm89may be true09:39
asm89just started working with sf2, i think it's quite nice actually09:39
asm89but for the OO port09:39
asm89isn't working with doctrine2 something that would be nice?09:40
darynasm89, better run.  I see jreese coming with a club!09:40
daryn:)09:40
* asm89 hides09:40
jreese /rage09:41
asm89but seriously, it is considered?09:41
darynasm89, depends on what you mean by considered...09:41
darynyes we thought about it09:41
asm89but now mantis is doing it itself?09:42
darynnot likely to use it.09:42
asm89mapping database data to oop models?09:42
*** Joins: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net)09:42
*** Parts: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net) ()09:42
dhx1asm89: I think the approach generally agreed upon is to keep things simple... no need to try and convert data from the database into OO classes09:43
asm89ah ok09:43
*** Quits: sgimeno (~chatzilla@163.117.206.10) (Quit: ChatZilla 0.9.87 [Firefox 4.0.1/20110413222027])09:43
jreeseasm89: in all seriousness, ORM systems like Doctrine2 are great in theory, and work really well for simple data models, but they break down really quickly when you start having complex data models, or doing complex queries, at which point you have to write your own SQL anyways...09:43
asm89true09:44
dhx1asm89: we have the Bug object already... which is really only useful for creating a new bug, reading a single bug at a time, etc09:44
dhx1asm89: as soon as you want to perform actions on multiple bugs (generating stats, etc)... the Bug object is mostly useless overhead09:44
asm89the point is ofcourse that you can't really map an object data structure to a sql database09:44
asm89but i didn't realise that the object model for mantis would be so complex09:45
dhx1asm89: more to the point however... most frameworks force you into using certain template systems, application structures, etc09:45
asm89agreed09:45
jreeseit's a lot easier to use a framework when you're starting from scratch09:45
asm89true09:46
jreeseit's a completely different beast to shoehorn a framework into a mature legacy app like mantis09:46
asm89im not saying mantis should :P09:46
asm89im porting a legacy app to sf2 + doctrine2 now myself09:46
asm89and i think doctrine2 does a nice job of adding OOP flavor to php, which can benefit a lot of codebases09:47
asm89but i also understand the KISS principles of mantis09:47
dhx1asm89: we do have a lot of complex things that are going on behind the scenes... trees (hierarchical project structures), graph theory (relationships between bugs), etc09:47
dhx1asm89: if you want to choose which bugs to display on the bug list page the process can be quite complex09:48
asm89hm09:48
dhx1filters that need to use tree traversal over projects, access control considerations (RBAC or otherwise), etc09:48
dhx1all balanced against performance and ease of maintenance09:49
asm89yes09:50
asm89but you can combine the two?09:50
asm89doing the complex query, but still binding the result to a Bug object in the end?09:50
asm89to work with on the output side09:50
dhx1you're right... there are many valid opportunities to use OO code09:50
dhx1and a lot of the work done so far (especially by Daryn) on the output side has been investigating PHPTAL (XML based templating engine)09:51
darynwhich, btw, likes using objects for output09:52
dhx1yep09:52
asm89ah cool :)09:52
asm89i just really like mantis (have been using it for quite a while now)09:52
asm89but didn't really dug into the sourcecode09:53
dhx1it's my choice as well because it's so customisable and simple09:53
asm89one time to add a ticket status :P but i thought it was quite horrible ;)09:53
dhx1the source code tends to be old and messy in some places09:53
darynasm89, word to the wise....Don't!  (dig into the source code)09:53
dhx1but for the most part it's quite simple... just messy in places09:53
dhx1avoid filters :)09:53
asm89hehe09:54
daryni'd say it's messy all over09:54
asm89is the phptal template engine a final decision?09:54
darynmmm...is anything ever final?09:54
dhx1it's understandable though... you can read most MantisBT source files and know what it's trying to do fairly easily09:54
dhx1of course there is a lot of work to do...09:54
dhx1but it's better than a lot of source code I've read which is overly abstract and OO-ified for the sake of OO (not clarity or any other reason)09:55
asm89true09:55
asm89any place where i can view the work done?09:59
*** Quits: kirillka (~Miranda@195.242.142.17) (Quit: kirillka)10:00
dhx1asm89: check daryn's branches on Github (forks of the mantisbt branch)10:00
darynasm89,  Well, it's a moving target but here is the refactoring branch https://github.com/daryn/mantisbt/tree/refactor10:01
dhx1I think he still has PHPTAL tests up there10:01
daryni haven significant changes that are not commited and periodically rewrite history on that branch10:01
dhx1at the moment it's just testing... (I think)10:01
darynbut you can get a general idea of where i'm headed10:01
darynmost of the other branches can be ignored...i should probably just delete them10:02
asm89cool, ill check it out10:03
asm89Development area for MantisBT contributions. History WILL get rewritten!10:04
asm89:P10:04
dhx1:)10:04
asm89autoloader aye?10:05
asm89:p10:05
dhx1yeah, faster method10:06
asm89        $t_class_name = str_replace( '\\', '/', $p_class_name );10:08
asm89and in all the other places DIRECTORY_SEPARATOR10:08
asm89:P10:08
dhx1not sure on that one10:08
dhx1haven't seen that bit of the code myself10:08
dhx1in many cases DIRECTORY_SEPARATOR is not needed (even when dealing with files)10:09
asm89https://github.com/daryn/mantisbt/blob/refactor/application/Autoloader.php10:09
* asm89 hides10:09
asm89:P10:09
asm89but the application directory is the "new" code10:10
darynasm89, those are going away10:10
asm89oh?10:11
darynthe DIRECTORY_SEPARATOR stuff10:11
*** Joins: tmckeown (~Adium@64.191.211.43)10:11
darynlike i said there are a lot of changes in my local branch that i need to finish and push10:11
daryni guess that would be the one place the DIRECTORY_SEPARATOR is needed and not using :D10:12
darynoops10:12
asm89ah ok10:13
asm89ill check another time :)10:13
asm89cool to see this irc channel is active though!10:13
darynkinda like a volcano10:13
darynlong dormant periods with brief explosions10:14
asm89hehe10:14
dhx1hah10:14
dhx1it hasn't always been that way though10:14
*** Parts: tmckeown (~Adium@64.191.211.43) ()10:15
dhx1it's typically been a project where people have contributed to MantisBT as part of their paid work10:15
dhx1for instance, they need new features in the bug tracker they're using within their organisation10:15
dhx1and by pure luck they happen to have a clue about software development and come here to contribute :)10:16
asm89hehe10:18
asm89i saw the bidding on tickets some while ago10:18
asm89how's that going?10:18
dhx1people offering money to complete tickets?10:18
asm89yea?10:19
jreesedhx1: that's the only reason the plugin system and source integration exist :P10:19
dhx1jreese: and a lot of other code too :P10:19
asm89haha10:19
asm89ok10:19
jreesetags, category inheritence...10:19
dhx1asm89: yeah... I think it gets overlooked because a) it's risky, b) it generally underpays ?10:20
dhx1I did have someone offer me a bounty before (which I didn't notice when fixing the issue)10:20
asm89haha10:20
dhx1but for $10-20... it's not really worth the hassle... and in my case, I wasn't looking for money on small bounties :)10:20
dhx1nice gesture however10:20
asm89yea10:21
dhx1I can't think of any projects where small bounties for bug fixes work?10:21
dhx1most projects tend to sponsor a student for a few months to fix as many bugs as possible10:22
darynmy time to read the ticket is worth more than the bounty10:22
dhx1or sponsor someone to add a big new feature, etc10:22
dhx1daryn: haha10:22
jreeseI would love to see Mantis participate is GSOC, but it means somebody has to put in a decent amount of work to a) create a proposal, and b) mentor and/or watch over the studentsif it gets approved10:22
dhx1yeah... and you'd need a student who knows what they're doing10:23
dhx1so they don't spend 3 months working on something that'll never get merged10:23
darynhm...that's an oxymoron10:23
dhx1(this is partly a supervisor responsibility too)10:23
darynhence the term, 'student'10:24
daryn:)10:24
dhx1you'd need to handhold the student for 5-6hrs/week I'd guess?10:24
dhx1:)10:24
dhx1I like the idea of it... just saying that it'd probably require the supervisor to dedicate a lot of time10:24
jreesedaryn: I would have known what I was doing for GSOC when I was a student, hell, all of my major contributions to mantis happened for the job I had while in uni10:24
dhx1ditto for me :)10:25
daryni was just funnin, guys...don't get all defensive on me10:25
dhx1and I guess GSoC is competitive enough that the poor programmers miss out?10:25
dhx1hah :)10:25
jreesedaryn: http://xkcd.com/386/10:25
darynheh10:26
asm89gsoc apply time is over right?10:27
jreeseI believe so10:27
jreeseI think they've already picked students for the projects too10:27
darynfwiw, i don't claim to know what i'm doing10:27
jreesedaryn: that's because you're not perfect and infallible like I am10:27
darynwhich is probably why i've refactored refactor several times10:28
jreeselol10:28
dhx1I'd be interested in helping for future proposals to join GSoC10:28
dhx1:)10:28
dhx1would MantisBT stand a chance?10:28
asm89lol daryn10:29
darynjreese, i know better...i've found some of your bugs10:29
daryn:P10:29
* daryn thinks maybe that could be worded better...10:29
dhx1:)10:30
jreesethose were all intentional ;)10:30
asm89haha10:30
dhx1not everyone can write perfect code... *wink* :P10:30
jreeseI was just testing how good of a team I was working with10:30
darynoh right...that was my training10:30
jreesecongrats, you passed10:31
daryngee thanks...do I get a diploma?10:31
jreeseno, open source projects can't afford that10:32
dhx1has anyone noticed that paulr is very good at finding bugs in your patches... but not his own? :P10:32
darynha!10:32
jreeseI noticed that about 2 years ago10:32
*** Joins: tmckeown (~Adium@64.191.211.43)10:32
darynone change to that sentence...10:32
jreesehe's also really good about reverting things that aren't his10:33
dhx1most of them seem related to using Windows and not having tools like grep readily available to mass search for functions that need removing, updating, etc when the API changes10:33
dhx1I'm being serious... he's probably found the most bugs out of anyone here in my patches :)10:33
darynpaur is very good at imagining bugs in your patches but not his own10:34
darynhehe10:34
jreesehe's also really good at finding "bugs" that are really just disagreements10:34
darynhe does like to point out other's errors10:34
jreeseI imagine it's because he's british10:35
dhx1daryn: see here... John pretending that his bugs are actually just "disagreements" :)10:35
darynthere you go imagining things too10:35
darynhehe10:36
jreeseit's not just my cmmits that I'm talking about10:36
dhx1maybe not "disagreements"... rather, "tests for Daryn"10:37
daryni know he's done the same for me.  but in his defense he has also found real bugs10:37
dhx1right, it's also the ability to press 'o' on the keyboard10:37
dhx1(grasping for material now)10:38
jreeseI think what bugged me the most was that he arbitrarily reverted rombert's commit that added a bunch of test bootstrapping and automation, just because he's supposedly been working on something for a different tool that he uses on mantisforge, and so many months later still haven't seen any mention of what he's done...10:39
jreeseesp bugged because there wasn't anything wrong or bad about what rombert commited, just that paulr didn't like the tool it used...10:41
darynyes i can't wait to see if he reverts my commit with all the oo changes...10:41
*** Parts: Rixie (~Rixie@188.177.20.182) ()10:42
darynprobably because he hasn't commited his db changes yet ( disregarding the fact that I've integrated them into my commit )10:42
*** Quits: cerf (~quassel@mcl71-3-78-241-52-239.fbx.proxad.net) (Remote host closed the connection)10:54
asm89im leaving now11:06
asm89till next time!11:06
jreesecheers11:06
*** Quits: asm89 (~asm89@5ED18FE0.cm-7-2c.dynamic.ziggo.nl) (Quit: bye!)11:06
*** Quits: giallu (~giallu@fedora/giallu) (Ping timeout: 260 seconds)11:38
*** Joins: giallu (~giallu@fedora/giallu)11:50
*** Quits: giallu (~giallu@fedora/giallu) (Read error: Operation timed out)12:06
*** Quits: wolfman2_ (~wolfman20@rrcs-70-63-208-211.midsouth.biz.rr.com) (Remote host closed the connection)12:39
*** Joins: cerf (~quassel@mcl71-3-78-241-52-239.fbx.proxad.net)13:46
*** Quits: cerf (~quassel@mcl71-3-78-241-52-239.fbx.proxad.net) (Remote host closed the connection)13:46
*** Quits: JonMarkGo (~Jon@pool-96-232-112-232.nycmny.fios.verizon.net) ()14:09
*** Joins: paulr_ (~IceChat09@2001:470:9310:aaaa:1d88:81c7:3d1c:8f7f)14:11
*** Joins: wolfman2000 (~wolfman20@rrcs-70-63-208-211.midsouth.biz.rr.com)15:21
*** Joins: giallu (~giallu@88.149.248.104)16:04
*** Quits: giallu (~giallu@88.149.248.104) (Changing host)16:04
*** Joins: giallu (~giallu@fedora/giallu)16:04
*** Joins: asm89 (~asm89@dhcp-077-251-157-210.chello.nl)17:15
*** Quits: asm89 (~asm89@dhcp-077-251-157-210.chello.nl) (Client Quit)17:17
*** Joins: asm89 (~asm89@dhcp-077-251-157-210.chello.nl)17:20
*** Quits: tmckeown (~Adium@64.191.211.43) (Ping timeout: 276 seconds)18:02
*** Quits: daryn (~daryn@h158.249.190.173.static.ip.windstream.net) (Quit: Ex-Chat)18:12
*** Quits: asm89 (~asm89@dhcp-077-251-157-210.chello.nl) (Quit: bye!)18:13
*** Joins: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net)18:18
*** Parts: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net) ()18:22
*** Quits: scribe9343423 (~scribe934@static.96.23.63.178.clients.your-server.de) (Remote host closed the connection)20:00
*** Joins: scribe9343423 (~scribe934@static.96.23.63.178.clients.your-server.de)20:00
*** Quits: paulr_ (~IceChat09@2001:470:9310:aaaa:1d88:81c7:3d1c:8f7f) (Quit: Copywight 2010 Elmer Fudd. All wights wesewved.)20:09
*** Quits: micahg (~micahg@ubuntu/member/micahg) (Read error: Operation timed out)20:18
*** Joins: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net)22:11
*** Parts: tmckeown (~Adium@pool-108-56-179-231.washdc.fios.verizon.net) ()22:11

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!